II. Does Timing Matter? A Real Options Experiment to Farmers’ Investment and
5.3 Test of H2 ‘learning effect’ and H3 ‘farmer-specific effects’
To test H2 and H3, we run a Tobit regression for each treatment with the individual (dis)investment year of farmers as the dependent variable. A Tobit model (Tobin 1958) is used to estimate linear relationships between variables if the dependent variable is censored as is the case in our study. The time of (dis)investment could only be observed if it falls between zero and nine. The results are presented in Table 5.
In our experiment, each farmer repeated treatment A and treatment B 10 times, so that in each case, farmers had 10 times the option to (dis)invest or not to (dis)invest. Thus, the variable ‘repetition’ can take a value between 1 and 10. In treatment A, the estimated coefficient of the ‘repetition’ variable is significant and has a positive sign (p-value < 0.001), meaning that with each repetition of the investment treatment, farmers invest 0.169 years later. This implies that they learn from their experiences of previous investment decisions. Although participants approximate the ROA in later repetitions, their investment behaviour still does not exactly follow an optimal manner. This result confirms previous findings of Oprea et al. (2009). The estimated coefficient of the variable
‘repetition’ in treatment B is not significant at 5 per cent (p-value = 0.071). Farmers do not approximate the predictions of the ROA over time, but they also do not further deviate from the ROA benchmark. On this basis, we fail to reject H2 in terms of investment and we reject H2 in terms of disinvestment.
In the experiment, we examined the presence of an ‘order effect’. Farmers were faced with both treatments in a different order, so that some were first faced with the investment treatment and then with the disinvestment treatment or vice versa. According to Scheufele and Bennett (2013), repeated choice tasks may influence outcomes through order effects.
The estimated coefficient of the ‘order’ variable is significant in both treatments, which shows that farmers demonstrate different (dis)investment behaviour dependent on the order in which they are faced with the two treatments. However, it may also indicate a ‘learning effect’, meaning that farmers acquire routines for repetitive decisions at the beginning of the experiment and apply them to later decisions even if they are related to another treatment.
Table 5 Results of the Tobit regression of the individual (dis)investment year (N = 2700) Coefficient p-value Coefficient p-value
Constant 8.324
University degree (1: yes, 0: no) 0.133 (0.271)
Notes: Asterisk * and double asterisk ** denote variables significant at 5 and 1 per cent, respectively.
Standard errors are indicated in parentheses.
In treatment A, the estimated coefficients of the variables ‘risk attitude’, ‘age’, and
‘household size’ are significant and have a negative sign, while the variables ‘economic background in education’ and ‘farm size’ are significant and have a positive sign. That means that more risk averse farmers, older farmers and farmers with a larger number of family members invest earlier, whereas farmers with an economic background in education and with a larger amount of farmland invest later and therefore more in accordance with the ROA. In treatment B, the estimated coefficients of the variables ‘risk attitude’, ‘age’, and ‘gender’ are significant and have a negative sign, while the variables ‘economic background in education’ and ‘farm size’ are significant and have a positive sign. This implies that more risk averse farmers, older farmers, and male participants disinvest earlier
and therefore more in accordance with the ROA, whereas farmers with an economic background in education and who own a larger amount of farmland disinvest later. In contrast to the investigations of Adesina et al. (2000), Carey and Zilberman (2002), O’Brien et al. (2003), Gardebroek and Oude Lansink (2004), Seo et al. (2008), Pushkarskaya and Vedenov (2009), and Willebrands et al. (2012), the variables ‘farm income type’, ‘irrigation use’, ‘farm type’, ‘university degree’, and ‘farm performance’ do not appear to affect (dis)investment decisions significantly. The non-significance of the variable ‘irrigation use’ may indicate that our results are not considerably influenced by the framing of our experiment. The findings of the variables ‘risk attitude’ and ‘age’ in respect to disinvestment and the variable ‘farm size’ regarding to both treatments confirm our expectations (Foltz 2004; Pushkarskaya and Vedenov 2009; Savastano and Scandizzo 2009; Sandri et al. 2010). It is revealed that farmers with an economic background in education result in later (dis)investment timing. It may indicate that farmers who have better information through their economic background in education put a higher value on the option to wait and, for this reason, (dis)invest later than less informed farmers. In both treatments, risk is found to play a role in the decision to (dis)invest in irrigation technology. However, it is surprising that risk averse farmers invest earlier, which is contradictory to our expectation that higher levels of individual risk aversion lead to later investment (Viscusi et al. 2011). A possible explanation for this behaviour may be that more risk averse farmers consider irrigation as a risk management instrument and, therefore, invest earlier than the less risk averse farmers. It is interesting to note that older farmers invest earlier, which might be explained by the fact that the participating farmers were relatively young with an average age of 32.1 years. Based on the literature, older farmers are expected to be less eager to invest in new technology (Gardebroek and Oude Lansink 2004). As many studies find that women invest later than men (Jianakoplos and Bernasek 1998), interestingly, there is no significant effect of the variable ‘gender’.
Moreover, men were found to disinvest earlier than women, while farmers with a larger household size were found to invest earlier, which does not support the findings of previous studies (Lewellen et al. 1977; Justo and DeTienne 2008). Based on the overall results, we fail to reject H3.
6 Conclusions
A better understanding of farmers’ decision to (dis)invest in agricultural technology under uncertainty is crucial for gaining insight into the dynamics of adoption and abandonment behaviour, interpreting agricultural outcomes, and designing policies that effectively assist farmers. This study examined the (dis)investment behaviour of farmers under flexibility, uncertainty, and irreversibility, while trying to determine the underlying models of investment consistent with the observed decision behaviour during an experiment. The (dis)investment decisions were modeled as real options, which refer to the rights to acquire and to sell irrigation technology. The observed (dis)investment decisions were contrasted with normative benchmarks, which were derived from the NPV and the ROA.
Our findings were first that neither the NPV nor the ROA provided an exact prediction of farmers’ (dis)investment behaviour observed in the experiment. Farmers invested later than predicted by the NPV but earlier than suggested by the ROA. Regarding the disinvestment situation, farmers disinvested later than predicted by the NPV and even later than suggested by the ROA. The results also suggested that the ROA can predict actual (dis)investment decisions better than the NPV. Second, we found that farmers accumulated knowledge through repeated decision-making in investment situations, and hence approximated the predictions of the ROA, but did not further deviate from the ROA benchmark in disinvestment situations. Third, we found that certain socio-demographic and farm-specific variables affected the (dis)investment behaviour of farmers.
When interpreting the results, it is important to take into account that our experimental design is abstracted from reality and is considerably simpler than (dis)investment situations that would occur in an actual business setting. Participants may behave differently in the experimental situation than they do in a similar situation in the real world. Decision makers who are faced with real (dis)investment problems (e.g. technology adoption and abandonment) often have multiple objectives, and they require more time to prepare and to make these far-reaching decisions. An individual’s decision behaviour can also be affected by perceptions and beliefs based on available information and can be influenced by attitudes, motives and preferences (McFadden 1999). A common criticism of experiments has to do with whether experimental results are likely to provide reliable inferences outside the laboratory and can be extrapolated to the real world (Levitt and List 2007; Roe and Just 2009). This lack of external validity is considered to be the major weakness of laboratory experiments (Loewenstein 1999). Framing might help render a laboratory experiment more
realistic and, thereby, increase its external validity. Several studies discuss the relevance of framing effects on choices given the fact that decision makers might be more ‘attached’ to a project that is described in terms that are more familiar to them (Cronk and Wasielewski 2008; Patel and Fiet 2010). Actually, there is an intensive debate on the trade-off between the internal and external validity of economic experiments (Camerer, 2003; Guala, 2005).
However, there is a widespread consensus that the benefits of internal validity are more important than the lack of external validity if the experiments aim to test economic theories, as is the case in our study (see Schram 2005).
The general implication from this experimental analysis is that flexibility, uncertainty, and irreversibility play a role in farmers’ decision-making process to adopt and abandon irrigation technology. This is extremely relevant from a policy maker’s perspective. It highlights the danger of designing policy measures solely based on the NPV given that this approach is not individually sufficient in order to explain (dis)investment decisions. The NPV fails to address the role of sunk costs, temporal flexibility, and uncertainty in the farmer’s decision-making process. However, it also is not sufficient to solely focus on the ROA when designing appropriate policy measures, since demographic and socio-economic factors also play a role. Policies that allow farmers to be more certain of future returns or practices that can reduce the uncertainty might encourage a more responsive (dis)investment strategy, regardless of the decision makers’ risk attitude. This is particularly relevant if there are public and environmental benefits arising from the adoption of new technologies, such as water-saving technologies and technologies to reduce land salinity. Policy measures, such as subsidies, might improve the adoption of more efficient water-saving practices and technologies. However, this also implies that under uncertainty, the rates of subsidy, which are required to encourage faster uptake of water-saving technologies are likely to be higher than those indicated by the NPV criterion.
In addition, it is a challenge for policy makers to consider the effects of certain socio-demographic and farm-specific characteristics on (dis)investment behaviour in the course of the current socio-demographic change in many countries. One example, which might be relevant for (dis)investment decisions, is the ageing of the population. Ageing may change decision makers’ (dis)investment behaviour. An understanding of the (dis)investment decisions taken by farmers is therefore important for the formulation of adequate forecasts and policy recommendations in the agricultural sector.
The experimental investigation of real options settings is still in its early stages. In this regard, further research is required for a better understanding of what exactly drives an individual’s decision-making in (dis)investment situations and to predict this behaviour in the future. It is possible that potential drivers of psychological inertia also play a role when explaining (dis)investment behaviour. Furthermore, it would be interesting to reveal the heuristics, which participants apply in order to make (dis)investment decisions. Another interesting research avenue would be the testing of whether farmers in developing countries show a similar (dis)investment behaviour as farmers in developed countries.
References
Abel, A.B. and Eberly, J.C. (1994). A unified model of investment under uncertainty, The American Economic Review 84, 1369–1384.
Adesina, A.A., Mbila, D., Nkamleu, G.B. and Endamana, D. (2000). Econometric analysis of the determinants of adoption of alley farming by farmers in the forest zone of southwest Cameroon, Agriculture, Ecosystems and Environment 80, 255–265.
Baerenklau, K.A. (2005). Toward an understanding of technology adoption: risk, learning, and neighborhood effects, Land Economics 81, 1–19.
Brennan, D. (2007). Policy interventions to promote the adoption of water saving sprinkler systems: the case of lettuce on the Gnangara Mound, The Australian Journal of Agricultural and Resource Economics 51, 323–341.
Camerer, C.F. (2003). Behavioral Game Theory: Experiments in Strategic Interaction.
Princeton University Press, Princeton, USA.
Camerer, C.F. and Hogarth, R.M. (1999). The effects of financial incentives in experiments: a review and capital-labor-production framework, Journal of Risk and Uncertainty 19, 7–42.
Cameron, L.A. (1999). The importance of learning in the adoption of high-yielding variety seeds, American Journal of Agricultural Economics, 81(1), 83–94.
Carey, J.M. and Zilberman, D. (2002). A model of investment under uncertainty: modern irrigation technology and emerging markets in water, American Journal of Agricultural Economics 84, 171–183.
Charness, G., Gneezy, U. and Kuhn, M.A. (2012). Experimental methods: Between-subject and within-subject design, Journal of Economic Behavior and Organization 81, 1–8.
Cronk, L. and Wasielewski, H. (2008). An unfamiliar social norm rapidly produces framing effects in an economic game, Journal of Evolutionary Psychology 6, 283–
308.
DeTienne, D. and Cardon, M.S. (2006). Entrepreneurial exit strategies: the impact of human capital, Frontiers of Entrepreneurship Research 26, 1–14.
Dixit, A. and Pindyck, R. (1994). Investment under Uncertainty. Princeton University Press, Princeton, USA.
Foltz, J.D. (2004). Entry, exit, and farm size: assessing an experiment in dairy price policy, American Journal of Agricultural Economics 86, 594–604.
Frey, G.E., Mercer, D.E., Cubbage, F.W. and Abt, R.C. (2013). A real options model to assess the role of flexibility in forestry and agroforestry adoption and disadoption in the Lower Mississippi Alluvial Valley, Agricultural Economics 44, 73–91.
Gardebroek, C. and Oude Lansink, A.G.J.M. (2004). Farm-specific adjustment costs in Dutch pig farming, Journal of Agricultural Economics 55, 3–24.
Guala, F. (2005). The Methodology of Experimental Economics. Cambridge University Press, Cambridge, UK.
Hill, R.V. (2010). Investment and abandonment behavior of rural households: an empirical investigation, American Journal of Agricultural Economics 92, 1065–1086.
Holt, C.A. and Laury, S.K. (2002). Risk aversion and incentive effects, The American Economic Review 92, 1644–1655.
Huettel, S., Musshoff, O. and Odening, M. (2010). Investment reluctance: irreversibility or imperfect capital markets?, European Review of Agricultural Economics 37, 51–76.
Jianakoplos, N.A. and Bernasek, A. (1998). Are women more risk averse?, Economic Inquiry 36, 620–630.
Justo, R. and DeTienne, D.R. (2008). Gender, family situation and the exit event:
reassessing the opportunity-costs of business ownership, IE Business School Working Paper GE8-108-1.
Kaplan, E.L. and Meier, P. (1958). Nonparametric estimation from incomplete observations, Journal of the American Statistical Association 53, 457–481.
Kiefer, N.M. (1988). Economic duration data and hazard functions, Journal of Economic Literature 26, 646–679.
Knight, J., Weir, S. and Woldehanna, T. (2003). The role of education in facilitating risk-taking and innovation in agriculture, Journal of Development Studies 39, 1–22.
Koundouri, P., Nauges, C. and Tzouvelekas, V. (2006). Technology adoption under production uncertainty: theory and application to irrigation technology, American Journal of Agricultural Economics 88, 657–670.
Levitt, S.D. and List, J.A. (2007). What do laboratory experiments measuring social preferences reveal about the real world?, The Journal of Economic Perspectives 21, 153–174.
Lewellen, W.G., Lease, R.C. and Schlarbaum, G.G. (1977). Patterns of investment strategy and behavior among individual investors, The Journal of Business 50, 296–333.
Loewenstein, G. (1999). Experimental economics from the vantage-point of behavioural economics, The Economic Journal 109, 25–34.
Longstaff, F.A. and Schwartz, E.S. (2001). Valuing American options by simulation: a simple least-squares approach, Review of Financial Studies 14, 113–148.
Luong, Q.V. and Tauer, L.W. (2006). A real options analysis of coffee planting in Vietnam, Agricultural Economics 35, 49–57.
Maart-Noelck, S.C. and Musshoff, O. (2013). Investing today or tomorrow? An experimental approach to farmers’ decision behaviour, Journal of Agricultural Economics 64, 295–318.
McFadden, D. (1999). Rationality for Economists?, Journal of Risk and Uncertainty 1-3, 73–105.
Musshoff, O. and Hirschauer, N. (2008). Adoption of organic farming in Germany and Austria: an integrative dynamic investment perspective, Agricultural Economics 39, 135–145.
Musshoff, O., Odening, M., Schade, C., Maart-Noelck, S.C. and Sandri, S. (2012). Inertia in disinvestment decisions: experimental evidence, European Review of Agricultural Economics, 40, 463–485.
O’Brien, J.P., Folta, T.B. and Johnson, D.R. (2003). A real options perspective on entrepreneurial entry in the face of uncertainty, Managerial and Decision Economics 24, 515–533.
Oprea, R., Friedman, D. and Anderson, S.T. (2009). Learning to wait: a laboratory investigation, The Review of Economic Studies 76, 1103–1124.
Patel, P.C. and Fiet, J.O., 2010. Enhancing the internal validity of entrepreneurship experiments by assessing treatment effects at multiple levels across multiple trials, Journal of Economic Behavior and Organization 76, 127–140.
Pushkarskaya, H. and Vedenov, D. (2009). Farming exit decision by age group: analysis of tobacco buyout impact in Kentucky, Journal of Agricultural and Applied Economics 41, 653–662.
Richards, T.J. and Green, G.P. (2003). Economic hysteresis in variety selection, Journal of Agricultural and Applied Economics, 35(1), 1–14.
Roe, B.E. and Just, D.R. (2009). Internal and external validity in economics research:
tradeoffs between experiments, field experiments, natural experiments, and field data, American Journal of Agricultural Economics 91, 1266–1271.
Sandri, S., Schade, C., Musshoff, O. and Odening, M. (2010). Holding on for too long? An experimental study on inertia in entrepreneurs’ and non-entrepreneurs’ disinvestment choices, Journal of Economic Behavior and Organization 76, 30–44.
Savastano, S. and Scandizzo, P.L. (2009). Optimal farm size in an uncertain land market:
the case of Kyrgyz Republic, Agricultural Economics 40, 745–758.
Scheufele, G. and Bennett, J. (2013). Effects of alternative elicitation formats in discrete choice experiments, The Australian Journal of Agricultural and Resource Economics 57, 214–233.
Schram, A. (2005). Artificiality: the tension between internal and external validity in economic experiments, Journal of Economic Methodology 12, 225–237.
Seo, S., Segarra, E., Mitchell, P.D. and Leatham, D.J. (2008). Irrigation technology adoption and its implication for water conservation in the Texas High Plains: a real options approach, Agricultural Economics 38, 47–55.
Tobin, J. (1958). Estimation of relationships for limited dependent variables, Econometrica 26, 24–36.
Trigeorgis, L. (1996). Real Options. MIT Press, Cambridge, UK.
Viscusi, W.K., Phillips, O.R. and Kroll, S. (2011). Risky investment decisions: how are individuals influenced by their groups?, Journal of Risk and Uncertainty 43, 81–106.
Willebrands, D., Lammers, J. and Hartog, J. (2012). A successful businessman is not a gambler. Risk attitude and business performance among small enterprises in Nigeria, Journal of Economic Psychology 33, 342–354.
Winter-Nelson, A. and Amegbeto, K. (1998). Option values to conservation and agricultural price policy: application to terrace construction in Kenya, American Journal of Agricultural Economics 80, 409–418.
Yavas, A. and Sirmans, C.F. (2005). Real options: experimental evidence, The Journal of Real Estate Finance and Economics 31, 27–52.
Appendix S1: Experimental instructions
Translation from German; Instructions for (dis)investment in irrigation technology General information
[…] The game consists of four parts and would require approximately 45 minutes of your time. Please read the following instructions carefully as your earnings in the experiment will depend on your decisions. Of course, your data will be treated as confidential and will be analyzed anonymously. […]
In each game, you should try to collect as many Euros (€) as possible because your potential earnings are proportional to the number of Euros (€) you collect during the game.
Besides an expense allowance of 10 €, each participant has three times the chance to receive a bonus if he/she completes the entire game. In the first and second parts of the game, one player is randomly selected and is given 100 € cash per 2,500 € achieved in a randomly selected round. The selected players for both parts will, therefore, receive between 270 € and 1,900 € as well as between 0 € and 1,900 €, respectively. In the third part of the game, again one player is randomly selected and is given a cash bonus of between 30 € and 1,155 €. In total, around 125 farmers can participate in the game. They will be informed via e-mail by December 10, 2011 if they receive one of the three cash bonuses in addition to the expense allowance. The earnings can be paid out or transferred to an account specified by the player selected.
Good luck!
Please note that submitted decisions during the game cannot be changed.
First part (instruction: treatment A (investment))
The game consists of various repetitions of one game with an equal basic structure.
Imagine that you as a farmer have liquid assets of 10,000 € at your disposal. Due to the ongoing phenomenon of global warming, the climate changes, which has an increasingly noticeable impact on agricultural production. Therefore, you are considering purchasing an irrigation system. In the time frame between 0 and 9 years, you can invest in an irrigation system only once. You can decide within the next 10 years:
- to immediately invest in an irrigation system
- to wait and see the development of the gross margins that can potentially be achieved (up to 10 years) and to invest in an irrigation system later
- or not to invest in an irrigation system.
The liquid assets you dispose of in your account in a given year will yield an interest rate
The liquid assets you dispose of in your account in a given year will yield an interest rate